Maker Pro
Maker Pro

Future: 0603 versus 0402 parts

M

MassiveProng

Jan 1, 1970
0
Do you?

If so, please enlighten us.

Thanks,
Rich
**** you, retard! This from the twit that claims to not read my
posts.

As I said, dipshit, I already did. All you have to do is look back
in the thread, or even perform a google search, if you aren't so
goddamned alcoholically retarded that you can do that.
 
M

MassiveProng

Jan 1, 1970
0
It's all
in designing the footprints and solder paste stencil correctly.


Exactly. Too much solder, and... whoop there it is!

BTW, the stencils are usually done by the fab house. If you design
your own, you are the exception. That is, however, the best way to
manage the amount of paste that gets pulled.
 
M

MassiveProng

Jan 1, 1970
0
Even some consumer stuff even to this day
still uses through hole. Look at a typical TV main board or a stereo board.
Or cheaper power supplies. Many of those are single sided even.
 
M

MassiveProng

Jan 1, 1970
0
I know what you mean - after working with a bunch of 0402's, I sometimes
confuse 0603's with them!


So, there is nothing wrong with using larger parts - the suppliers keep
pushing smaller is better (I think to push sales of new assembly equipment,
and to cover the investment costs of the production of smaller components).
I see no benefit in going smaller unless there is no room. I do however see
lots of problems in going smaller. Mostly reliability concerns and
production costs (especially hidden ones - ie change this resistor for this
mod now takes 5x longer and may be less reliable).


That's nasty!

I once resoldered my entire old laptop's motherboard (P2 233) littered with
0201's with a 1/8" chisle tip soldering iron! This fixed the intermittant
problem when you twisted it a bit.
It probably had nothing to do with any of the "soldering" you
performed and more to do with the interconnects between all the parts
you dis-assembled, and re-assembled. D'oh!
 
J

Jeff L

Jan 1, 1970
0
Spehro Pefhany said:
Yes, that's the hefty 0402. There's also a metric 0402 which is 0.4 x
0.2mm (01005), which is getting pretty dern small. The regular 0402s
can easily disappear into a blob of solder or go flying into
nothingness. With massive 0805s you generally should not need to buy
any extras even for hand assembly, but I always allow for a few extra
of the tiny ones to account for evaporation.


What! no extra's? I just saved 0.03 buy not ordering a few extra parts, but
now have to wait 2-3 days and $8 shipping, and spend 20 minutes of cost for
ordering / receiving / paperwork / book keeping for that little resistor
that went flying across the room!

Rule of thumb, always have at least 10 extra passive components and 1 or 2
extra of everything else, unless it's not a problem to have the project
delayed. 10% extra of the small parts / passives is generally something we
try to stick to, along with a few of the bigger / expensive components.

http://www.toplinecomponents.com/drawings/pdf/capacitors/SC01005.pdf


Or if you're myopic, just get close enough without searing your
corneas with splashes of flux.


Best regards,
Spehro Pefhany
http://www.speff.com
 
J

Jeff L

Jan 1, 1970
0
Hawker said:
I tend to use 0603 and only go 0402 if I have to.
There are a few reasons for this.
From an assembly standpoint, as another poster said 0402 is a knee in
the manufactures abilities. Also 0402 uses a different feeder for most
P&P machines than 1210-0603 so often CMs don't have as many feeders for
0402.


This is something that is of some importance - there are not that many extra
8x2 feeders kicking around from years of production like the common 8x4
feeders. Extra feeders are expensive, typically in the $1,000 plus range for
new ones (72 mm feeders are about $7,000), and each different part type
needs a feeder. 8x2 feeders are very hard to find used, and if used, they
may be worn out / need calibration, since 0402's and smaller need more
precise pickup points.

An example is we have around 400 - 500 8x4 feeders, about 75 8x2 feeders,
and about 200 of the other larger sizes.
From an electrical standpoint for caps there has been some recent
research showing that in the smaller geometries (0402 and smaller) the
caps do not hold there rated capacitance as well when they have a DC
bias on them (as in used for bypass caps). I am still learning about
this but apparently the actual capacitance value can go way down under
DC bias for smaller geometries and this is not as prevalent for 0603 and
larger.

This is a big problem when you push dielectrics to their max! Y type caps
suffer horrendously from this.

Also 0306, 0508 and 0612 caps are way better at reducing stray inductance
anyway!


<snip>
 
M

MassiveProng

Jan 1, 1970
0
Rule of thumb, always have at least 10 extra passive components and 1 or 2
extra of everything else, unless it's not a problem to have the project
delayed. 10% extra of the small parts / passives is generally something we
try to stick to, along with a few of the bigger / expensive components.
I have been fighting to get my company to include attrition in their
procedures.

I wish they could grasp the concept that it is actually cheaper to
have it, and perhaps not ever need it, than to have to rush to get it
in short quantity later.
 
D

DJ Delorie

Jan 1, 1970
0
Jeff L said:
I see no benefit in going smaller unless there is no room.

Well, my current project is space limited, which pushed me into 0603
and tssop sizes, but even then I didn't need to go to 0402s. The 0603
is big enough for my talon to grab, and I can get pretty much
everything (caps, resistors, LCL filters, LEDs, ferrites) all in the
one size.

FYI my current project puts an embedded linux system, five MCUs, four
bidirectional EMI filters with line boosters, 14 24VAC line switches,
27 LEDs, and a switching power supply... on a 3.5 by 5.5 board.
http://www.delorie.com/house/furnace/pcb2/
I do however see lots of problems in going smaller. Mostly
reliability concerns and production costs (especially hidden ones -
ie change this resistor for this mod now takes 5x longer and may be
less reliable).

Well, I just do my own boards one at a time, so most of that doesn't
affect me. But, I've not yet successfully removed and *reused* an
01005 chip. They just vanish in the solder. 0402 is probably as
small as I'll ever go for my boards, the 0201 can be done but they're
small enough to not be worth the hassle in general.
That's nasty!

No, it's challenging! http://www.delorie.com/pcb/smd-challenge/

The feedback seems to be split evenly between "Yeah, I could do it"
and "Hey, where'd that cap go?"
 
J

Jeff L

Jan 1, 1970
0
If enough care is taken, yes, it can work well. Most of my boards are
also rather mixed. The extreme was a board with 0603 on there but also a
DIP because the LM331 doesn't come any other way.

It's common to use dip parts and SMT!


I know but Digikey price ratios for reel qties are usually a pretty good
indicator when weighing one technology against another.

I used similar tactics when designing in components. I really like it when
companies include a unit price @ say 1 k on their website - it really helps
when making that part decision. TI has won a few design ins from me, just
from this alone.
Sometimes they
even beat distributor "pre-haggling" prices.

We typically only get that with new suppliers! they quickly learn, or we
don't do business with them. It's typically a problem with the smaller
suppliers, or the brokers that are trying become distributors.

There are a few cases however where in "low volumes" Digikey can't be beat.
Digikey excels at the prototype and low volume stuff, but for anything of
any volume, it's expensive.


 
J

Jeff L

Jan 1, 1970
0
Joerg said:
Thanks for sharing this, Frank. Those are the tricks of the trade that
really help. Nobody might know why but they do.

It relays to the surface tension of the solder - the rounded corners reduce
the area that the solder's surface tension can pull from, while still
keeping the pad at a reasonable length for placement tolerances and solder
volume.
 
J

Jeff L

Jan 1, 1970
0
Another thing that relates to tombstoning is the age, type, and
condition of your paste, as well as the time delay between pulling the
paste and reflow. 6 Hrs. is the typical industry maximum.

We found very little problems with solder paste age, at least with the type
of paste we use. There was no corrilation to tombstoning.


Doing a process study on it, we found that the only thing we found was old
solder paste (in 6 + months of cold storage) has little effect, and 6+
months of room temp storage gives more mid chip solder balls, as the solder
beads oxidize some and don't coalesce as well. Leaving the paste on the
stencil for long periods (a few days) in the tests also did not cause any
problems other then the paste drying out some (which kneading it brought it
back to usable), although it could cause some moisture absorption which can
cause some problems with solder voiding in BGA spheres. We even tested a few
boards by printing, placing parts, and letting them sit in an uncontrolled
environment (an unused top shelf in production) in increments to up to 6
months after the first tests of a few weeks showed no significant changes.
What we found was the tackiness mostly disappeared and the paste hardened a
bit over the period. Solder balling increased over the 6 moth period, and
wetting was not quite as good, but acceptable. Fine pitch had no problems.
 
J

Jeff L

Jan 1, 1970
0
Joerg said:
Yeah, I just had that happen. They "improved" the via clearances by some
insanely large margin, splitting a voltage plane into four islands. Arrgh.




I'd expect good fab places to know that. Actually I have never seen much
in terms of tombstoning problems even with lots of 0402 on there. So far.

As mentioned in my other post, we did a bunch of tests on this, and it seems
that the paste has little to do with this, at least with the paste we have
chosen to use.

The biggest problem that paste condition causes is solder balls in no clean
scenarios (this has much more to do with pad designs, and the no clean does
not get washed, so the solder balls are stuck in the flux residue) as the
paste ages. There is special paste avaible that has a mixed alloy which
melts over a slightly larger range, other then the eutectic point of 183 deg
C, which helps aide in reducing tombstoning, but we never needed it.
 
F

Frank Raffaeli

Jan 1, 1970
0
J

Jeff L

Jan 1, 1970
0
MassiveProng said:
It probably had nothing to do with any of the "soldering" you
performed and more to do with the interconnects between all the parts
you dis-assembled, and re-assembled. D'oh!

No, it was a microfracture along one of the components soldering. If I
recall right, it was in the video area. The board failed in several places
before (one was a large 2220 style common mode choke where the power adaptor
sent power through - you could see where it was arcing along the crack! and,
no it was not from the power connector flexing the board in this location,
as it was a fair distance away from that). I also had the laptop a part a
few times earlier diagnosing a faulty charging circuit. This laptop had a
lot of use, and was very thin at the time and thus flimsy, causing the unit
/ board to flex if you for example picked it up from one corner. I needed
the laptop, and I could not afford replacement parts / a new one at the time
(poor student), so it was worth the long evening fixing it. I still have it
kicking around somewhere, but it needs another hard drive.
 
J

Jeff L

Jan 1, 1970
0
MassiveProng said:
Or cheaper power supplies. Many of those are single sided even.

The same with the TV's and stereos. Single sided boards are cheap!

Even our fairly new "Inverter" microwave (that blew a IGBT) uses a single
sided TH board. Strange magnetics in that one!
 
F

Frank Raffaeli

Jan 1, 1970
0
Surface tension & meniscus.

Have to make sure the board shop doesn't mess with you gerbers
though, as some do.

Another thing that relates to tombstoning is the age, type, and
condition of your paste, as well as the time delay between pulling the
paste and reflow. 6 Hrs. is the typical industry maximum.

This may have more to do with fillet (pronnounced "fill-it") angle and
uneven wetting. Which I know you have a firm grasp on. ;-)

Frank
 
J

Jeff L

Jan 1, 1970
0
DJ Delorie said:
Well, my current project is space limited, which pushed me into 0603
and tssop sizes, but even then I didn't need to go to 0402s. The 0603
is big enough for my talon to grab, and I can get pretty much
everything (caps, resistors, LCL filters, LEDs, ferrites) all in the
one size.

FYI my current project puts an embedded linux system, five MCUs, four
bidirectional EMI filters with line boosters, 14 24VAC line switches,
27 LEDs, and a switching power supply... on a 3.5 by 5.5 board.
http://www.delorie.com/house/furnace/pcb2/

Nice looking board!
Well, I just do my own boards one at a time, so most of that doesn't
affect me. But, I've not yet successfully removed and *reused* an
01005 chip. They just vanish in the solder.

Likely doable with hot air.
0402 is probably as
small as I'll ever go for my boards, the 0201 can be done but they're
small enough to not be worth the hassle in general.


No, it's challenging! http://www.delorie.com/pcb/smd-challenge/

The feedback seems to be split evenly between "Yeah, I could do it"
and "Hey, where'd that cap go?"

:)
 
J

Jeff L

Jan 1, 1970
0
John Larkin said:
If he has 0402 dimensions that work better than mine, I'd be stupid
not to ask. I'm not afraid to admit that I don't know everything.

Having mounted a lot of parts, there are pad designs that work much better
then others. They may all fit with in the IPC requirements, and some that
work better may not! IPC rules are generally very good "guidelines".
 
J

Joerg

Jan 1, 1970
0
Jeff said:
I know what you mean - after working with a bunch of 0402's, I sometimes
confuse 0603's with them!




So, there is nothing wrong with using larger parts - the suppliers keep
pushing smaller is better (I think to push sales of new assembly equipment,
and to cover the investment costs of the production of smaller components).
I see no benefit in going smaller unless there is no room. I do however see
lots of problems in going smaller. Mostly reliability concerns and
production costs (especially hidden ones - ie change this resistor for this
mod now takes 5x longer and may be less reliable).

I am wondering about long term problems in RoHS countries with all that
fine pitch. Whiskers and all that. I've asked in a German NG a couple
times but got no hard answers. However, a rumor buzzes around that there
might be a fine-pitch exemption coming. I guess that would give the
small parts biz a big boost if the exemption materializes.

[...]
 
J

Joerg

Jan 1, 1970
0
MassiveProng said:
Or cheaper power supplies. Many of those are single sided even.


And on phenolic no less. Almost any cheap remote, toy, TV, thermostat is
phenolic thru-hole. That is probably why TI brought out many of the new
MSP430Fxxx in DIP.
 
Top